Expression analysis of circulating miR-22, miR-122, miR-217 and miR-367 as promising biomarkers of acute lymphoblastic leukemia

The role of serum-based biomarkers such as microRNAs in cancer diagnosis has been extensively established. This study aimed to determine the expression levels of bioinformatically selected miRNAs and whether they can be used as biomarkers or a new therapeutic target in patients with acute lymphoblastic leukemia (ALL). The expression levels of serum miR-22, miR-122, miR-217, and miR-367 in 21 ALL patients and 21 healthy controls were measured using quantitative real-time PCR. The receiver operating characteristic (ROC) curve and the associated area under the curve (AUC) was used to assess candidate miRNAs’ diagnostic value as a biomarker. The results showed that miR-217 was markedly decreased in patients with ALL compared to controls. Moreover, miR-22, miR-122, and miR-367 were found to be upregulated. Furthermore, ROC analysis showed that serum miR-217 and miR-367 could differentiate ALL patients from healthy individuals, while miR-22 has approximate discriminatory power that requires further investigation. These results provide promising preliminary evidence that circulating miR-217 and miR-367 could be considered potent diagnostic biomarkers and therapeutic goals in this disease.


Introduction
Acute lymphoblastic leukemia (ALL) is a neoplastic abnormality of the lymphoid lineage known primarily as a childhood disease, with 75% occurrence in children under the age of 10 years [1]. The underlying mechanisms of leukemia are numerous genetic mutations that lead to chromosomal abnormalities and malignant cell division and proliferation composed of immature and dysfunctional lymphocytes [2]. Although inappropriate gene function and altered gene expression patterns are a hallmark of ALL, but not enough to cause leukemia and it is most likely caused by interaction between exogenous and endogenous factors, genetic or some hereditary factors, and chance [3]. Careful assessment of prognosis, risk factors, and rapid diagnosis is essential for ALL management and successful treatment [1]. Overall, Tumor-based tissue extraction for diagnosing and follow-up is difficult and painful for patients, and in some cases, it is not repeatable. Therefore, there is less desire to use diagnostic strategies based totally on bone marrow or tumor biopsy to monitor the treatment of patients. New alternatives that overcome these negative aspects are increasing today, such as liquid biopsy [4,5]. The role of bio-fluids-based biomarkers and liquid biopsy in cancer diagnosis has been extensively established due to their non-invasive protocols and ease of acquisition [6]. Liquid biopsies such as cell-free DNA and miRNAs are also used in the diagnosis, prognosis, and follow the treatment response of hematological malignancies and may also minimize the need for invasive strategies such as bone marrow biopsy in the future. Liquid biopsy can be helpful in acute leukemia, monitoring initial treatment response and minimal residual monitoring [7]. Since the discovery of the first miRNA (a small non-coding singlestranded RNA molecule) called lin-4 (Lineage-4) in 1993 [8], numerous studies have shown that inappropriate miRNA expression is associated with many human diseases, including cancer [9,10]. Interestingly, they can be as "oncomiRs (oncogenic miRNAs)" or "tumor suppressor miRNAs" and affect the significant carcinogenic pathways that are often involved in apoptosis, cell proliferation, angiogenesis, metastasis, and drug resistance [9]. In addition, miRNAs could be considered a biomarker of disease due to their tiny structure and stability under various conditions such as boiling, multiple freeze-thaw cycles, and pH changes [10], as well as secretion into extracellular bio-fluids including serum and plasma [11], saliva [12], urine, seminal fluid, breast milk, cerebrospinal fluid, bronchial lavage and also tears [13].
It has recently been shown that miRNA expression profiles vary across different types of leukemia and are associated with survival and prognosis [9,14,15]. In this study, according to the genes and signaling pathways involved in the ALL, four related miRNAs (miR-22, miR-122, miR-217, and miR-367) were selected using reliable predictive databases and-previous studies [16][17][18][19]. Despite therapeutic advances, there is still frequent treatment failure in adult ALL, with a low long-term cure rate. In addition, despite the high rate of ALL treatment in childhood, relapsed and refractory diseases are still one of the common causes of death in this group [20]. Hence, there is an indispensable need to look at new diagnostic and treatment goals. Thus, this study aimed to assess the potential impact of miRNAs as serum biomarkers or new therapeutic targets in patients with ALL.

Subjects
The study protocol was approved by the Medical Ethics Committee of Bushehr University of Medical Sciences, Bushehr, Iran (no. IR.BPUMS.REC.1397.059, approved October 22, 2019) regarding the declaration of Helsinki. The samples and clinical information of patients were collected according to the guidelines of the relevant ethics committee with prior informed consent from all participants. The present cross-sectional study included 21 patients, most aged under 15, newly diagnosed with ALL before any chemotherapy, and 21 healthy individuals without underlying disease matched for gender and age. Moreover, samples of the same ALL patients were enrolled in the study after routine treatment, which is risk-based multiagent combination chemotherapeutic protocol consists of steroids, alkylating agents, antimetabolites, microtubule-destabilizing agents, nucleoside analogues, anthracyclines and hydrolyzing enzymes. Whole blood samples of participants were collected in gel separation vacutainer tube. Due to the importance of immediate separation of whole blood into the serum to maintain RNA stability, the samples were processed within one hour. They were centrifuged at 2000×g for 10 min at 4 °C, followed by 15,000×g to remove cell debris. Then the serum supernatant was aliquot to 1.5-mL cryotubes, then stored at − 80 °C until further use.

Bioinformatics analysis
In the present study, the possible role of miRNAs in ALL was investigated. First, genes and signaling pathways involved in ALL pathogenesis were surveyed based on literature and database [21,22]. Then according to these paths, the most critical miRNAs were selected for further analyses. Two primary database, miRBase and miRWalk, were used to get more information about miRNA features and interactions between miRNAs and genes associated with the ALL, respectively [23,24]. The DIANA TOOLS and KEGG (Kyoto Encyclopedia of Genes and Genomes) pathway analysis were employed to predicts the candidate miRNAs targets of a particular function in diseases or pathways, based on three algorithms: Tarbase, MicroT-CDS, and Targetscan [25][26][27]. Subsequently, the desired microRNAs were selected focusing on KEGG pathways and based on the results of the predictive sites.

miRNA isolation and cDNA synthesis
For circulating miRNA analysis, direct miRNA quantification from serum was performed using chemical denaturing buffer composed of 2.5% tween-20, 50 mM Tris-HCl, and 1 mM EDTA, followed by heat denaturation to release circulating miRNAs from both microvesicles and exosomes [28,29]. First, samples were centrifuged at 15,000 rpm for 10 min, then 10 µL of each sample's supernatant was incubated with 10 µL of tween-20 buffer for 10 min at 70 °C in a Dry Bath (Dolphin Co., India). After that, the samples were immediately incubated on ice for 5 min for Tween-20 settling and then centrifuged at 12,000 rpm for 10 min, and the supernatant was used for further cDNA synthesis steps. The poly(A) method was used for cDNA synthesis by reverse transcription kit (BonYakhteh Co., Iran) as the manufacturer's instructions recommended. In this method, a poly(A) tail is added to the 3' end of each mature miRNA by poly(A) polymerase, followed by a specific adaptor oligo(dT) primer for the reverse transcription [30]. The purity and quantity of cDNA was assessed spectrophotometrically as 260/280 nm and 260/230 nm ratios, using NanoDrop DS-11 (DeNovix Inc., Wilmington, DE USA).

Quantitative real-time PCR (qRT-PCR)
The qRT-PCR was performed in duplicate measurements using SYBR Green Master Mix (Yektatajhiz CO., Iran) using miRNA-specific forward primers and universal reverse primer in 96-well plates with an ABI StepOnePlus RT-PCR system (Applied Biosystems, USA). 0.5 µL of the BON reverse primer (BN-0011.17) and each miRNA-specific primer were used for each amplification reaction. Table 1 shows the forward primer sequences developed by the company (BonYakhteh Co., Iran) and their calculated melting temperatures.
The temperature and schedule were set as one primary denaturation at 95 °C for 2 min followed by 45 cycles of amplification at 95 °C for 5 s, and finally, 60 °C for 30 s to confirm amplification specificity. Moreover, no-template control (NTC) and no-RT control (NRC) were used to monitor PCR product specificity and contamination with genomic DNA, respectively. Furthermore, melting curve analyses and agarose gel electrophoresis of the PCR products were used to ensure the target gene amplification.

Statistical analysis
Data quality control was performed using a boxplot to identify and modify outliers for the miRNAs expression levels in controls and patients. The relative changes in target miRNAs expression was calculated using the comparative Ct method, known as the 2 −ΔΔCt method 1 [31]. According to the studies and the manufacturer's recommendation, miR-17-5p was used as an internal control to normalize the cycle thresholds (Ct) values, considering the stable and invariant expression between all samples [14,[32][33][34]. The Shapiro-Wilk test evaluated the normality of the data distribution. According to the results of data set distribution, the unpaired T-test or Mann-Whitney test were performed to compare the differences in miRNAs expression between the healthy controls and patients, and also paired T-test or Wilcoxon for comparison between the patient's group before and after treatment, using statistical package for social sciences (SPSS) 16 (SPSS Inc., Chicago, IL, USA) and GraphPad Prism 7 (GraphPad Prism Software, Inc. San Diego CA, USA). Furthermore, the receiver operating characteristic (ROC) curve and the associated area under the curve (AUC) was used to determine the diagnostic value of candidate miRNAs' as a biomarker. A p-value less than 0.05 was considered statistically significant for each test.

Characteristic of patients
The study population included 21 patients with newly diagnosed ALL, consisting of 13 men (61.9%) and 8 women (38.1%), and 21 age-and gender-matched healthy controls without underlying disease. Supplementary Table S1 summarizes the demographic and laboratory characteristics of the patients at the time of diagnosis. According to the results, expression of miR-22 was significantly higher in the adult group, while there was no significant difference in the relative expression levels of other candidate miRNAs between men and women, children and adults, and the ALL subtypes. Moreover, no statistically significant correlation was observed between the expression of miRNAs with age, platelets count, white and red blood cell count, and hemoglobin.

Differential expression of serum miRNAs in newly diagnosed ALL patients
Levels of serum miRNAs were assessed by qPCR to evaluate their role in the diagnosis of ALL. As shown in Fig. 1A-C, relative serum miR-22 and miR-367 expression increased significantly and up-regulated in the newly diagnosed ALL group compared to control (P = 0.01 and 0.002, respectively). Although the serum miR-122 levels showed increased expression, it was not statistically significant (P = 0.06). Moreover, we found that the relative levels of serum miR-217 expression were markedly decreased, and it was downregulated in the newly diagnosed ALL group compared to the controls (P = 0.00, Fig. 1D).

The changes in serum miRNAs expression in ALL patients following treatment
One month after the first routine treatment of the patients, their blood samples were examined to evaluate the changes in the expression level of the miRNAs with the mentioned methods. The results demonstrated that the expression levels of miR-22, miR-122, and miR-367 in the follow-up groups were decreased compared to the first examination (P = 0.05, P = 0.78, and P = 0.005, respectively, Fig. 1E-G). Interestingly, the expression level of miR-217 was significantly higher in the follow-up group than in the preliminary assessment (P = 0.01, Fig. 1H).
The diagnostic performance of candidate miRNAs for ALL patients ROC curve analysis was assessed to determine the specificity and sensitivity of each miRNA as a possible biomarker in the diagnosis and follow-up of ALL patients. To distinguish the newly diagnosed ALL group from the control, an AUC for miR-22, miR-122, miR-217 and miR-367 was found to be 0.7 with a sensitivity of 61% and a specificity of 71% (P = 0.03, 95% CI 0.69-0.94), 0.62 with a sensitivity of 33% and a specificity of 100% (P = 0.17, 95% CI 0.45-0.79), 0.81 with a sensitivity of 95% and a specificity of 61% (P = 0, 95% CI 0.69-0.94) and 0.8 with a sensitivity of 80% and a specificity of 81% (P = 0.001, 95% CI 0.65-0.94); respectively ( Fig. 2A-D).

Logistic regression results
The data of qPCR were used for binary logistic regression. According to the results of ROC curves, miR-217 and miR-367 were selected to simultaneously evaluate the role of biomarkers due to their appropriate sensitivity and specificity and higher Yuden index in newly diagnosed ALL patients. The mean logit (P) value of the ALL group was found to be 0.83 and significantly higher than that of the control group, 0.17 (P < 0.0001). The predicted values of logistic regression analysis showed that the combined ROC analysis of these two miRNA revealed an increased AUC value of 0.95, with a sensitivity of 95%, a specificity of 90%, and a Yuden index of 0.85 (P = 0, 95% CI 0.88-1.01; Fig. 2J-K). A logit (P) of 0.50 was used for optimal cutoff by which ALL and controls were distinguished (Table 3).

Discussion
MiRNAs are currently well defined as small regulatory molecules whose significant role is in the post-transcriptional regulation of their target genes and play crucial roles in physiological and pathological cellular processes [36]. To date, the expression profile and dysregulation of microRNAs has been studied in more than a dozen disease, particularly various types of human cancer [37][38][39]. Since the aberrant expression of serum miRNAs in cancer was first described [40], their role in tumorigenesis has been highly regarded by researchers. Importantly, miRNAs are effective biomarkers for classifying and tracing the source tissue of unknown primary and poorly differentiated cancers [41,42]. These findings emphasize miRNAs' diagnostic value in cancers and suggest their therapeutic role.
To the best of our knowledge, this study is the first report on the assessment of serum miR-22, miR-122, miR-217, and miR-367 in ALL and the use of combined miR-217 and miR-367 as biomarkers in this disease. Specifically, we found that the levels of miR-22, miR-122, and 367 were upregulated in ALL patients compared with controls, and additionally, miR-217 was markedly reduced. The analysis of ROC data revealed that serum miR-217 and miR-367 alone could be effective as a biomarker for early detection of ALL, while miR-22 has approximate discriminatory power that requires further investigation. Moreover, the combination of miR-217 and miR-367 further increased the diagnostic value and could distinguish newly diagnosed ALL patients from healthy controls.
Previous studies have examined the different roles of these miRNAs in various cancers. They have been described as double-faced molecules that can act as both a tumor suppressor and an oncogene. Bar N et al. demonstrated that overexpression of miR-22 can affect the tumor suppressor, i.e., PTEN and its downstream pathways PI3/AKT, thus inhibiting AKT proliferative and anti-apoptotic activity [43]. Another study by Song et al. reported that miR-22 is a crucial regulator for hematopoiesis maintenance, and it was found to be upregulated in myelodysplastic syndrome and leukemia, associated with a poor prognosis. In this study, the TET-2 gene was mentioned as a target of miR-22, which its deletion is involved in hematologic malignancy [44]. Given that PTEN activity is lost in many types of cancer, leading to increased cell survival and cell cycle progression [45], as well as the relationship between TET-2 and hematopoietic cell proliferation [46], it seems that miR-22 upregulation may lead to ALL due to its effect on these genes. Conversely, miR-22 was mentioned as a tumor suppressor in a study on acute myeloid leukemia (AML), which affects the MYC pathways [47]. Due to the various functions of miRNAs in different tissues, a discrepancy in the results is conceivable. In addition, we found that the miR-22 expression level was not significantly related to patient gender and laboratory data but was higher in the adult group. This could be related to the predominant subtype of adult ALL, T-ALL, which has multiple mutations such as NOTCH1 that lead to impaired differentiation and proliferation [48]. In the study of Saccomani et al., miR-22 decreased expression was reported, indicating the functional role of this miRNA in this subtype of ALL in coordination with NOTCH1 [49]. Therefore, the association with age may be due to an interaction with ALL subtypes. In another considerable study, a relationship between miR-22 and sex, platelet count, and AML subtypes was not reported, whereas a significant association was found with white blood cells and cytogenetics [50]. Similar to the results of our study, Manfe et al. mentioned miR-122 as an oncogene that enhances the anti-apoptotic pathway of AKT/P53 and represses chemotherapy-induced cytotoxicity in cutaneous T-cell lymphoma [17]. Fornari et al. also evaluated the association between miR-122 and apoptosis and identified cyclin G1 as a target of miR-122 [51]. By contrast, Yang et al. demonstrated that the expression of miR-122 in AML patients was significantly lower than in the healthy group and may function as a tumor suppressor. Moreover, they showed that the induction of miR-122 could inhibit cell proliferation in both HL60 and K562 cell lines of AML compared to the control cell line [52]. Consistent with the data of Zhang et al. on gastric cancer, in the present study, no significant relationship was observed between miR-122 expression levels and laboratory or demographic findings of patients [53]. Interestingly, MEF2D-rearranged ALL is associated with evasion of miRNA-mediated degradation. In this regard, a study has demonstrated that mutation in a miR-122 binding site increased MEF2D expression, leading to B-cell differentiation disorders by inhibiting PAX5 activity, a critical regulator of B-cell differentiation [54]. Hence, miR-122 may be a potential therapeutic target in this type of disease.
According to the present study results, a marked downregulation of miR-217 was observed that confirmed the tumor suppressor action of miR-217. We found a significant difference in the expression level of miR-217 in the post-treatment group compared to before treatment, which may be due to the effects of the treatment process. Our results are consistent with a previous study that demonstrated that the expression of miR-217 is decreased in colorectal cancer with the effect of the MAPK pathway, which regulates tumor growth and increases apoptosis [55]. Other studies have identified AKT3 [56], Runx2 [57], E2F3 [58], and KRAS [59] as targets of miR-217. The interaction of miR-217 and these genes

3
appear to be involved in pathogenicity in ALL. By contrast, in the study by Yebenes et al., miR-217 was mentioned as an oncogene that increases specifically in the germinal center of lymph nodes. According to this study, miR-217 reduces the expression of genes that respond to DNA damage and repair and stabilizes the BCL-6 expression [60]. Although the ROC analysis in a present study showed that miR-217 is favorable as a biomarker for diagnosing and following up the patients with ALL, it could be more potent in diagnosing. Consistent with these results, in a study by Yan et al., ROC analysis showed that miR-217 could effectively discriminate AML patients from healthy individuals. Also, the expression of miR-217 was remarkably increased in patients who reached complete remission after treatment, which is similar to our study's results, indicating the potential role of miR-217 in leukemogenesis [18].
In the previous study, miR-367 was identified as a regulator of embryonic stem cell self-renewality and pluripotency, which induces the trait of being stemness in the cancer cell and contributes to the progression and spread of cancer [61]. Consistent with the present study's data, Kaid et al. showed that miR-367 plays an oncogenic role in medulloblastoma cells by affecting the OCT4 transcription factor, leading to poor survival and poor prognosis [62]. In another similar study, miR-367 was mentioned as an oncogene that targets  tumor suppressor KLF4. It could be blocked at both mRNA and protein levels and its subsequent pathways, BAX and caspase-3, which suppress apoptosis [63]. Most studies indicated that miR-367 acts primarily as an oncogene. However, contrary to the present study's findings, it was shown that the expression of miR-367 was significantly reduced in gastric cancer tissues [64] and oral carcinoma cells [65].
According to the results of ROC analysis and based on the sensitivity, specificity, and Yuden index, we found that among the studied miRNAs, miR-217 and miR-367 have a more desirable ability to diagnose ALL, and a combination of these two miRNAs could reach a higher sensitivity and specificity. The limitations of our study were; limited access to newly diagnosed ALL patients, small sample sizes, and lack of access to patients' follow-up samples after the treatment process, which was due to the patient's death or referral to out-of-reach medical centers.

Conclusion
In conclusion, our results suggest that miR-217 and miR-367 could be potential new biomarkers for ALL diagnosing and a new therapeutic target. Further comprehensive studies are warranted to confirm these results and investigate the possible mechanisms and role of genes associated with these miRNAs in ALL.